1. Field of the Invention
The present invention relates to a method of recording/reproducing data on a recording medium.
2. Description of the Related Art
A method has been conventionally known, in which original data is converted into a code such that more than one "0" is interposed between bits "1" such as MFM code and RLL code (1,7), the converted code is recorded in a recording medium, a timing of a zero-cross of a reproductive signal or a second order differential signal is detected on the recording medium to obtain a bit "1", and the data is reproduced. In this data reproducing method, as shown (a) in FIG. 1, the original data is converted into the code in such a manner that more than one "0" is interposed between bits "1" a mark "1" is recorded by switching presence and absence thereof, timing with a bit "1" as shown in (b) and (c) of FIG. 1, and a timing at which a reproductive signal shown in (d) of FIG. 1 or a second order differential signal shown in (f) of FIG. 1 crosses a predetermined threshold value or a zero level is detected to obtain the bit "1" so as to reproduce the original data.
However, in the conventional recording/reproducing method, when a short mark such as "01010" is reproduced, the timing of the bit "1" indicated by .largecircle. mark in (d) and (f) of FIG. 1 is offset before or after from an original position. So, the conventional method is defective in that many errors are liable to be caused in such a portion, and therefore fail to improve a recording density. That is, in case of the reproductive signal, the position of the bit "1" is offset in a direction where there are more bits "0". In case of the second order differential signal, the bit "1" is offset in a direction where there are less bits "0" (i.e. one bit "0"). An interference between marks as it is called thus arises. The interference between marks becomes more and more noticeable as the recording density is higher and the mark is shorter. As a result, when the higher recording density is aimed for, errors are liable to be caused in the reproduced data.